Methionine Adenosyltransferase and Transmethylation in Fetal and Neonatal Lung of the Human, Monkey, and Rabbit

METHIONINE ADENOSYLTRANSFERASE AND TRANSMETHYLATlON 545

20. Yavin, E., and Menkes. J. H.: Glyceride metabolism in cultured cells dissociated Institute of Science. Rehovot, Israel. from rat cerebral cortex. J. Neurochem., 21: 901 (1973). 28. The current address of Dr. A. H. Nash is: 9615 Brighton Way, Beverly Hills. 2 1. New England Nuclear, Boston. Mass. Calif. 22. Pentex Miles Lab. Kanakee. Ill. 29. This work was supported in part by United States Public Health Service 23. Reheis Chemical Co., Chicago, Ill. 1-POI-HD055 15-01, I-ROI-HD0928 1-01, and I-T32-GM07015-01. 24. Analtech, Canoga Park, Calif. 30. Requests for reprints should be addressed to: A. Milunsky, MB.Ch., M.R.C.P., 25. Supelco, Inc.. Bellefonte, Pa. D.C.H., Genetics Laboratory, Eunice Kennedy Shriver Center, 200 Trapelo 26. Consent to study the children was provided by the parents in all cases. Road, Waltham, Massachusetts 02154 (USA). 27. The current address of Dr. E. Yavin is: Biochemistry Department, The Weizman 3 1. Accepted for publication December 3, 1975.

Pediat. Res. 10: 545-550 (1976) Fetus transmethylation lung transsulfuration methionine adenosyltransferase

Methionine Adenosyltransferase and Transmethylation in Fetal and Neonatal Lung of the Human, Monkey, and Rabbit

HANS J. STERNOWSKY, NlELS C. R. RAIHA, AND GERALD GAULL"" Department of Pediatric Research, New York State Institute for Basic Research in Mental Retardation, Staten Island, New York; Department of Pediatrics and Clinical Genetics Center, Mount Sinai School of Medicine of the City University of New York, New York, USA; Departments of Medical Chemistry and Obstetrics and Gynecology, University of Helsinki, Helsinki, Finland

Extract extrauterine survival is possible in each species suggests that this enzyme may be important in the synthesis of surface-active lecithin Optimal conditions for the assay of methionine adenosyltrans- and/or in methylation reactions related to detoxification of the ferase in crude extracts of human fetal lung were determined. constituents in the blood perfusing the breathing lung. Maximal activity was obtained with 36 mM ATP, 20 mM L-methionine, 240 mM Mg++,and 160 mM K+. The pH optimum was 6.2-6.4, which is the same as that for adult human lung but lower than that for human liver. In human fetal iung, there was an S-Adenosylmethionine, a major donor of methyl groups in increase in specific activity of methionine adenosyltransferase with mammalian tissues, is formed from L-methionine and ATP in the increasing gestational age (r = 0.87; P < 0.01) up to 25 weeks of presence of Mg++ and K+ by rnethionine adenosyltransferase gestation, after which time no fetal specimens were obtained. The (ATP:L-methionine S-adenosyltransferase, EC. 2.5.1.6) (2). Bio- specific activity of 5-methyltetrahydrofolate-homocysteine me- synthesis of S-adennsylrnethionine is also the first step on the thyltransferase of human fetal lung was 2.51 & 0.88 nmol/mg transsulfuration pathway, which transfers the sulfur atom from proteinlhr, which was higher (P < 0.001) than the activity found in L-methionine to the 3-carbon skeleton of L-serine to form newborn lung (0.14 =t 0.01). Activity of serine hydroxymethyltrans- L-cysteine. In human fetal liver and brain, but not in kidney, the ferase and of betaine-homocysteine methyltransferase was absent transsulfuration pathway (see Fig. 1) is incomplete (I 2, 24, 28) as a from human fetai lung. Activity of cystathionine synthase was result of the absence of cystathionase (L-cystathionine cysteine- absent from fetal, neonatal, and mature human lung. Activity of lyase (dearn~nating),EC. 4.2.1. l j, the enzyme which cleaves cystathionase in fetal and newborn human lung was present only in cystathionine to cysteine and a-ketobutyrate. trace amounts. In rhesus monkey lung, beginning 15 days before In the human, the transsulfuration pathway apparently becomes term, the activity of methionine adenosyltransferase increased active after birth (30). In the mature human, 90% of the 6-fold to reach a maximum before term (165 days), and during the methionine sulfur goes down this pathway (26); therefore, we first weeks of life the activity gradually diminished. 5-Methyltet- examined some of the enzymes of transsulfuration and related rahydrofolate-homocysteine methyltransferase activity in fetal reactions in human fetal tissues. Investigation of the remethylation (100-145 days) monkey lung was higher (6.57 rt 0.95 nmol/mg of homocysteine to methionine by the vitamin BIZ-dependent proteinlhr) than in newborn lungs (1.91 + 0.97) (P < 0.001). In 5-methyltetrahydrofolate-homocysteine methyltransferase (5- fetal rabbit lung, the activity of methionine adenosyltransferase de- methyltetrahydropteroyl-L-glutamate:L-homocysteine S-methyl- creased 2.5-fold during the last third of pregnancy, whereas a 2-fold transferase, EC. 2.1.1.13) demonstrated that the specific activity of increase occurred during the first 48 hr after term birth. this enzyme was much greater in fetal human liver and brain than in mature human liver and brain (14). The alternative enzyme for Speculation remethylation of homocysteine to methionine, betaine-homocysteine methyltransferase (EC. 2.1.1.5), was much less active (14). In The increasing specific activity of methionine adenosyltransferase addition, the apparent K, for 5-methyltetrahydrofolate-homocys- in the lung of the human, monkey, and rabbit around the time when teine methyltransferase is of the order of M, whereas that of 546 STERNOWSKY, R

cystathionine synthase, the enzyme which competes for the survival after premature birth of man and of the rhesus monkey substrate, homocysteine, is of the order of M (7). Further- (16-19). Others consider only the so-called choline pathway of more, the specific activity of cystathionine synthase also is reduced significance in fetal lung (5). The controversy in this field has been in human fetal liver and brain (12). Thus, 5-methyltetrahydrofo- reviewed recently (38). It also is possible that biologic methylation late-homocysteine methyltransferase would be expected to com- would be of importance in the detoxifiction (6) of a number of Pete effectively with cystathionine synthase for the free homocys- compounds in the lung. teine, thus forming a methionine cycle (Fig. 1). It was of interest, therefore, to examine the activity of some of The methionine cycle, the existence of which has been appreci- the enzymes of the methionine cycle and transsulfuration at times ated in mature tissues (7, 9), was postulated to be especially active when there might be an increased need for methylation. We find in human fetal liver and brain (10, 11). An active methionine evidence of increased activity of the methionine cycle in the lung cycle would conserve the sulfur of homocysteine at the price of during fetal life and evidence that the specific activity of methio- making cyst(e)ine essential and would facilitate the biosynthetic nine adenosyltransferase in lung increases at the time of gestation reactions related to this cycle. The fact that cystathionase is when extrauterine life becomes possible for man, monkey, and virtually absent would, according to this hypothesis, maximize the rabbit. conservation of homocysteine sulfur, since none could be converted to cysteine. We have provided ancillary evidence in support of the METHODS AND MATERIALS biologic significance of increased activity of the methionine cycle in fetal liver. (I)There is greater incorporaton of [35S]methionine The lungs from previable human fetuses of 3-25 cni crown-rump into proteins of human fetal liver than into proteins of mature length, approximately 6-25 weeks of gestation, were obtained human liver (12). (2) There is virtually no incorporation of the immediately after legal abortion by hysterotomy. None of the sulfur of methionine into the sulfur of cyst(e)ine by perfused investigators participated in any way in the clinical decision for human fetal liver (12). (3) There is a greater synthesis of abortion. Gestational age was determined from the crown-rump polyamines from S-adenosylmethionine in fetal human liver than length nomogram of Tanimura et al. (36). The lungs were in mature human liver (28). (4) In fetal human liver, at the time removed, freed from any visible blood by washing the pleura with that activity of 5-methyltetrahydrofolate-homocysteine methyl- isotonic saline, and immersed in liquid nitrogen. No fetus was transferase is high, there is more rapid transfer of the P-carbon of observed to breathe after removal from the amniotic sac. Human serine into the de novo synthesis of DNA via thymidylate (31) than newborn lungs were obtained from premature and term human there is in adult human liver. newborns who died of causes other than respiratory distress or This cycle also seems to provide the potential for an increased pulmonary disease, as determined clinically. Biopsies of these enzymatic capacity for biologic methylation. Although the specific lungs were performed through high axillary intercostal incision not activity of methionine adenosyltransferase in human fetal liver later than 2 hr after death. After removal and saline washing, if during the second trimester is less than that of mature human liver, necessary, these lungs were immersed in liquid nitrogen. Human the specific activity of this enzyme in human fetal brain increases adult lung was obtained during postmortem examination up to 4 hr during the second trimester (12), and in the rat (3) it increases after death. Specimens from adults with obvious pulmonary greatly in the liver at the time of maximal fetal growth, i.e., late in disease were excluded from the study. After freezing in liquid gestation. The so-called methylation pathway for synthesis of the nitrogen, all samples were stored at -70' until analyzed (no surface-active phospholipid, phosphatidyl choline (lecithin), is longer than 2 weeks). considered by some investigators to be of importance in allowing Female rhesus monkeys, weighing 2-3 kg, were obtained pregnant from the wild. They were kept in the Laboratory for GLYCINE CH2THF+ dTMP+ DNA Experimental Surgery in Primates, Tuxedo, New York, until PROTEIN operation or delivery. The stage of pregnancy was determined SYNTHESIS roentgenographically and by palpation with an accuracy of i 5 2" days. Fetuses at various stages of gestation were obtained by cesarian section under Nembutal anesthesia. They were prevented ~METHIONINE from breathing, if necessary, by capping the head with a condom. Lungs were removed and kept frozen as described above. Lungs were obtained from newly born monkeys delivered vaginally at term. Newborns at various ages were decapitated and the lungs were removed, washed, and frozen. S-ADENOSYL- Pregnant New Zealand albino rabbits were obtained from METHlONlNE Perfection Breeders, Douglasville, Pa. The time of conception was known within 6 hr. The dams were decapitated at several stages of gestation, the uterus removed, and the lungs of the fetuses quickly removed and immersed in liquid nitrogen. Dissection was performed with the aid of a dissecting microscope, whenever neces- S-ADENOSYL- sary. Newborn rabbits were decapitated at various stages after HOMOCYSTEINE METHYLATION. DNA,RNA. PROTEIN. birth, and the lungs were removed as described. 4 LECITHIN L-[3-"C]Serine (48 mCi/mol) was obtained from Amersham/ CYSTATHIONINE Searle, Arlington, Ill.; [methyl-14C]betaine was obtained from New England Nuclear, Boston, Mass. All other reagents were of PALP @ I the highest purity available from Sigma Chemical Co., St. Louis, CYSTEINE Mo., and were used without purification. Fig. I. The methionine cycle and related biosynthetic pathways. Sample preparation was carried out at 0-4'. Fetal lungs were CH,THF: 5-methyltetrahydrofolate; THF: tetrahydrofolate; CH,THF: homogenized in a Potter-Elvehjem glass-glass homogenizer for 4 5,lO-methylenetetrahydrofolate;dTMP: thymidylate; B,,, vitamin B,, in min in equal parts (w/v) of Tris buffer, 0.2 M, pH 7.2. Mature cofactor form; PALP: pyridoxal phosphate; SAM: S-adenosylmethionine; lungs were first homogenized in an OmniMixer from Sorvall, Inc., methionine adenosy1transferase;Q). various specific methyltransferases; Norwalk, Conn., with two parts of buffer (w/v) and then again 0:S-adenosylhomocysteine hydrolase; @ 5-methyltetrahydrofolate- homogenized in a glass-glass homogenizer. Supernates obtained homocysteine methyltransferase; 0:cystathionine synthase; @: cystathio- after centrifugation at 30,000 x g for I hr were used for all assays. nase; 0:serine-hydroxyrnethyltransferase. Activity of methionine adenosyltransferase was determined by METHlONlNE ADENOSYLTRANSFERASE AND TRANSMETHYLATION 547 quantifying the amount of S-adenosylmethionine formed under HUMAN FETAL LUNG conditions saturating for substrates (13). Except where indicated, In 18 human lungs from first and second trimester fetuses there the following, in micromoles, were incubated at 37' for 120 min in was an increase in the specific activity of methionine adenosyl- a total volume of 0.5 ml: Tris-HCI buffer, pH 7.2, 60; KCI, 80; transferase with increasing crown-rump length (r = 0.87; P < MgCI*, 120; ATP, 18; L-methionine, 10. Appropriate dilutions of 0.01) (Fig. 2). In nine newborn human lungs from 9 hr to 25 days of the tissue extract were made with the same Tris buffer, and the age, methionine adenosyltransferase was 3.05 * 1.52 nmol/mg other compounds were added as neutral aqueous solutions. proteinlhr (Table 1). No dependence on postnatal age, weight at S-Adenosylmethionine was quantified with a Beckman model birth, weeks of gestation, or the time after death of autopsy 120C amino acid analyzer (13). Calculation of the peak area and (within the limit of 2 hr) was observed In two adult lungs the subsequent calculations were carried out with an Infotronics IOOOL activity was 1.3 and 1.6 nmol/mg proteinlhr, respectively. on-line computer. Cystathionine synthase and cystathionase were 5-Methyltetrahydrofolate-homocysteine methyltransferase in 16 determined by our previously described assays (13). We used our fetal human lungs was 2.51 * 0.88 nmol/mg proteinlhr (Table I). minor modifications (14) of the radioassays for the determination No correlation with crown-rump length was observed. This value of betaine-homocysteine methyltransferase (8). 5-methyltetrahy- was far higher than that found in neonatal and mature lung. It was drofolate-homocysteine methyltransferase (8), and serine-tetrahy- half the specific activity we reported for fetal liver, but twice that drofolate-methyltransferase (EC. 2.1.1.1) (37). All assays were of mature liver (14). Betaine-homocysteine methyltransferase was performed in duplicate; the duplicates did not differ more than determined in six fetal (second trimester) human lungs; only a 10%. A blank determination with extract boiled for 5 min before trace of activity was found. Cystathionine synthase was deter- incubation was performed for each sample. Protein was determined in lungs from 5 fetuses, 13 newborns, and 3 adults; only mined by the method of Lowry et al. (21). trace activity was found. Cystathionase was determined in lungs from 5 fetuses and 13 newborns; it was found to have a specific RESULTS activity approaching the lower limit of detection in our assay. In ASSAY CONDITIONS FOR METHlONlNE ADENOSYLTRANSFERASE contrast to neonatal liver, in which the specific activity of Optimal conditions for the assay of methionine adenosyltrans- cystathionase and cystathionine synthase increased after birth, ferase in crude extracts of human fetal lung were systematically there was no postnatal increase. Serine hydroxymethyltransferase determined. Concentrations used for both substrates were saturat- was determined in 15 fetal lungs; the specific activities approached ing; 36 mM ATP and 20 mM L-niethionine give maximal activity. the lower limit of detection. There was no activity without added ATP, whereas without methionine it was about 20% of the maximal activity. This is RHESUS MONKEY FETAL LUNG similar to rat and human liver (33, 34), in which even with partially Because of the uncertainties of interpretation when postmortem purified methionine adenosyltransferase there was a considerable human tissues are utilized, the lack of availability of third activity without added methionine, but none without added ATP. The activity of methionine adenosyltransferase in human fetal lung was proportional to the concentration of soluble protein when analyzed in the range between 0.8 and 6.8 mg protein/assay in piperazine-N,N1-bis(2-ethanesulfonic acid) (Pipes) buffer, 0.8 M at pH 7.0. At low protein concentrations, it was observed that Tris buffer, which was used for the actual determinations of specific activity, caused a slight inhibition resulting in a deviation from linearity. This has been observed previously, but not commented on, by Mudd et al. (22). In our studies 3.0 6.0 mg protein were used, which is in the linear range for both buffers. The activity of methionine adenosyltransferase in human fetal 1 - lung was proportional to the time of incubation at 37O, with an 5 10 15 20 25 initial time lag of 4 niin when incubated with Tris buffer. This time CROWN-RUMP LENGTH IN CM lag disappeared when the reaction mixture was incubated with Fig. 2. Development of specific activity of methionine adenosyltrans- Pipes buffer at the same pH, again suggesting a slight inhibition of ferase in human lung. SAM: S-adenosylmethionine. methionine adenosyltransferase by Tris. Tris buffer rapidly loses its buffering capacity below pH 7.0. Table I. Enzymatic activities in hunzan lung and liver' Therefore, we also used Pipes buffer for the determination of pH optima, after it was established that Pipes does not activate or 5-Methyltetrahy- inhibit methionine adenosyltransferase in the pH range over 7.0. In drofolate- the pH curves, the pH at several points was rechecked after Methionine homocysteine incubation, and it did not differ from the value before incubation adenosyltransferase methyltransferase by more than 0.02 pH units. The pH optimum in Pipes buffer for human fetal lung was 6.2-6.4. At pH 7.0, the fetal lung enzyme in Fetal lung 2 2.51 i 0.8g3 (16) vitro was 75% of its maximal activity. Ault human lung also had Newborn lung 3.05 i 1.52 (9) 0.14 * 0.01 (6) a pH optimum of 6.2-6.4. At pH 7.0 the adult lung enzyme in vi- Mature lung 1.30; 1.60 0.10; 0.05 tro had 85% of its maximal activity. Human fetal liver, like hu- Fetal liver' 26 + 3 (24) 4.70 i 0.16(31) man adult liver (33, has a pH optimum at 7.0 (Pipes buffer) with Mature liver' 86i 16(9) 1.30 i 0.20(17) a broad peak. The activity of methionine adenosyltransferase was measured with Tris buffer at pH 7.2 in all lung extracts, since ' All values are nanomoles of product formed per mg of soluble protein many of our experiments were already done when the pH opti- per hr + standard error. Numbers in parentheses indicate number of mum of lung was found to be lower than that of liver. Because of specimens studied. the small size of the samples and the difficulty in obtaining the Increase in specific activity with increasing crown-rump length (Fig. 2). original human samples, it was not possible to repeat the determi- P < 0.001 by Student unpaired (-test when compared with newborn nation for methionine adenosyltransferase in all of them. Thus, the lung. values reported consistently represent about 75% of the maximal ' Values are taken from References 12 and 14 and presented for ease of activity. comparison. trimester human fetuses and of normal human newborns, and because of the similarity of the human to the monkey with regard to the metabolism of the lung (15) and development of sulfur amino acid metabolism (32), we examined the development of these enzymes in monkey lung. Methionine adenosyltransferase was determined in the lungs of 21 monkeys: 12 fetuses from 100 days to 155 days of gestation and 9 infants from 1 hr to 1 year of age. Term in the monkey is about 165 days. Lungs from fetuses obtained between 100 and 145 days of gestation have relatively low specific activity, 1.36 nmol/mg proteinlhr A 0.68. No correlation with gestational age could be demonstrated (Table 2). Immediately before term, however, the specific activity increases almost 6-fold. During the first weeks of life it gradually diminishes over a period of 3 months to reach the specific activity found in adult lung. One fetus, delivered by cesarian section on da.v 142 of gestation, was DAYS OF GESTATION HOURS AFTER BIRTH accidentally allowed to breathe for about 5 min. In this lung, the Fig. 3. Development of specific activity of methionine adenosyltrans- specific activity of methionine adenosyltransferase was found to be ferase in rabbit lung. The number under each point is the number of litters about 3 times higher than that of fetuses of this age that had not examined at each age or the number of individual dams. See text for breathed. 5-Methyltetrahydrofolate-homocysteine methyltransfer- details. SAM: S-adenosylmethionine. ase activity was 6.57 + 0.95 nmol/mg protein/hr in fetal monkey lungs and 1.91 * 0.97 in lungs from newborns. This difference was statistically significant (P < 0.001). These values were similar to ferase than male rat liver (23), this is not true in the rabbit (20); those found in fetal and mature liver of this animal (Table 2). therefore, lungs from male rabbits were not studied.

RABBIT FETAL LUNG DISCUSSION A total of 44 litters, with an average of 8 fetuses, and I2 adult The present data give evidence of increased activity of the rabbit lungs were analyzed for activity of methionine adenosylmethionine cycle in fetal lung. Methionine adenosyltransferase transferase. The specific activity decreases from 5.14 A 0.70 activity is increased at a time when 5-methyltetrahydrofolate nmollmg protein/hr on the 18th day of pregnancy to 2.03 + 0.58 methyltransferase activity is also increased and when activities of nmol/mg proteinlhr on the last day of pregnancy (day 31). After cystathionine synthase and cystathionase are absent. The increase birth, the activity increases during the first 48 hr of extrauterine in the methyltransferase (with its apparent K, for homocysteine of life to a maximum of 4.13 + 0.63 nmol/mg proteinlhr (Fig. 3). M), combined with the virtual absence of both enzymes of Lungs from one litter were analyzed when the pups were 72 hr of transsulfuration, would assure maximal recycling of homocysteine age and were even higher, 5.9 nmol/mg proteinlhr. Mean specific sulfur to methionine sulfur and transfer of the methyl group of activity in the adult lung, as determined in 12 dams, was 1.49 * 5-methyltetrahydrofolate for resynthesis of S-adenosylmethionine. 0.10 nmol/mg proteinlhr, which was significantly lower than the Maximal methylation capability would then be assured. The mean specific activity at birth (P < 0.05). Although female rat question might be posed: Why should there be any increase in liver has a higher specific activity of methionine adenosyltrans- specific activity of methionine adenosyltransferase when it is known to be operating at substrate concentrations well below its Table 2. Enzymatic activities in rhesus rnonke,: lung and liver1 K, for methionine? However, if there should be an increased demand for S-adenosylmethionine at such times in development 5-Methyltetrahy- when the concentration of methionine in the tissues actually is Methionine drofolate- decreasing, an increase in total methionine adenosyltransferase adenosyltrans- homocysteine activity might be necessary to provide the enzymatic capacity ferase methyltransferase required to meet this increased demand. The mechanism for regulation of methionine adenosyltransferase activity appears to Fetal lung be complex (4, 34), suggesting that it may be involved in regulatory 100-145 days 6.57 1 0.952 functions and demonstrating that factors other than substrate 142 days, breathing 3.79 concentration are important. 158 days 3.34: 5.76 The increasing specific activity of methionine adenosyltransfer- Neonatal lung ase activity in the lung of the human, monkey, and rabbit around I hr the time when each species acquires the respiratory capability to 7 days survive is striking. A change in composition of the blood perfusing 9 days the lungs at this time might elicit such an increase, since methyla- 30 days tion reactions may be involved in the detoxification of both endog- I lodays enous and exogenous substances (6). It is also possible that the in- 168370 days crease in the enzymatic potential for methylation might be related Fetal liver3 to the respiratory function of the lung. (145160days of Human and monkey fetuses have the respiratory capability to gestation) survive premature birth by the dint of the ability to synthesize Mature liver3 surfactants prior to term. In both species, around the time that survival becomes possible (at about 28 weeks for the human; at ' All values are nmoles of product formed per mg of soluble protein per about 150-155 days for the monkey, with about 50% survival at hour standard error. Numbers in parentheses indicate number of 145 days (24)), the activity of methionine adenosyltransferase specimens studied. increases (Fig. 4). In the monkey, there is a perinatal peak of P < 0.001 by Student unpaired t-test when compared with values from activity which begins about 2 weeks before term, which is just the 5 days before to l I0 days after birth. time that Gluck and coworkers (15) have reported the lecithin to Values are taken from unpublished studies by J. A. Sturman and G. E. sphingomyelin ratio of the amniotic fluid in the monkey to rise. 17 Gaull (29) and are given for the purpose of comparison with lung. the human, there is a steady increase in specific activity during the METHlONlNE ADENOSYLTRANSFEIRASE AND TRANSMETHYLATION 549

methionine in vivo into choline and phosphorylcholine and proposed that S-adenosylmethionine is the methyl donor for the a 10 - I I I I methylation of ethanolamine, phosphorylethanolamine, and prob- . Term 5 I I ably CDP-ethanolamine, as well as of phosphatidylethanolamine. If their work is correct and if it applies to lung, then it is possible 0 LL that the increase in specific activity of methionine adenosyltrans- a Monkey 6 ferase, and, therefore, in availability of S-adenosylmethionine, is m - .E involved in synthesis of phosphatidylcholine by either or by both Z routes. a 4- U) SUMMARY -Y", 0 2- Methionine adenosyltransferase activity in lung is increased at a E time when 5-methyltetrahydrofolate methyltransferase activity is C also increased and when activities of cystathionine synthase and .I- 3 3 6 6 100 cystathionase are absent. Maximal enzymatic capability is found in the lung of the human, monkey, and rabbit around the time TIME OF PREGNANCY IN "/.OF TOTAL when each species acquires the respiratory capacity to survive.

Fig. 4. Comparison of development of specific activity of methionine REFERENCES AND NOTES adenosyltransferase in lung from human, monkey, and rabbit with regard I. Bremer. J., Figard, P. H., and Greenberg. D. M.: The biosynthesis of choline and to the time of pregnancy. SAM: S-adenosylmethionine. its relation to phospholipid metabolism. Biochim. Biophys. Acta, 43: 477 ( 1960). 2. Cantoni, G. L.: S-Adenosylmethlonine: A new intermediate formed enzymati- second trimester. Third trimester human fetuses and normal cally from L-methionine and adenosine trlphosphate. J. Biol. Chem., 204: 403 (1953). human neonates cannot be studied; however, the similarity of the 3. Chase. H. P., Volpe, J. J.. and Laster. L.: Transsulfuration in mammals: Fetal monkey to the human with regard to other aspects of sulfur and early development of methionine-activating enzyme and its relat~onto metabolism (32) suggests that the increase occurring during the hormonal influences. J. Clin. Invest., 47: 2009 (1968). 4. Chou, T.-C., and Talalay, P.: The mechanism of S-adenosyl-L-methionine second trimester probably continues into the third trimester. The synthesis by purified preparations of bakers' yeast. Biochemistry, 11: 1065 mean specific activity of lung methionine adenosyltransferase of (1972). the human neonate is below the specific activity late in the second 5. Farrell, P. M., Lundgren. D. W., and Adams, A. J.: Choline kinase and choline trimester but probably above that in mature lung. It also is phosphotransferase in developing fetal rat lung. Biochem. Biophys. Res. possible that the specific activity was lower in neonatal lung than in Commun.. 57: 696 (1974). 6. Feuer, G., Miller. D. R., Cooper. S. D., de la Egles~a,F. A., and Lumb, G.: The the fetal lung in our study because the neonates all had been sick influence of methyl groups on toxicity and drug metabolism. Int. J. Clin. prior to death or because the enzyme is unstable postmortem. Pharmacol. Ther. Tox~col.,7: 13 (1973-1). Zachman (40) also reported methionine adenosyltransferase activ- 7. Finkelstein, J. D.: Methionine metabolism in mammals. In: N. A. J. Carson and ity in lungs obtained postmortem from newly born infants, but he D. N. Raine: Inherited Disorders of Sulphur Metabolism (Churchill Living- stone, London, 1971). used a different assay, and it is not clear that his results are 8. Finkelstein, J. D., Kyle, W. E., and Harris. B. J.: Methionine metabolism in comparable with ours. One major discrepancy between his results mammals: Regulation of homocyste~nemethyltransferase in rat tissues. Arch. and ours is that he found methionine adenosyltransferase activity Biochem. Biophys.. 146: 84 (1971). in neonatal lung higher than in neonatal liver, whereas we found 9. Finkelstein, J. D., and Mudd, S. H.: Transsulfuration in mammals. The meth~onine-sparingeffect of cystine. J. Biol. Chem., 242: 873 (1967). liver at all stages of development to have a higher specific activity 10. Gaull. G. E.: Interrelationships of sulfur amino acids, folate and DNA in human than lung. No explanation of this discrepancy is apparent to us. brain development. F. Hommes and C. J. von den Berg: Symposium on Rabbits cannot breathe prior to I or 2 days before term. In the Development Biochemistry-Inborn Errors of Metabol~sm. (Academic Press. New York, 1973). rabbit fetus, in contrast to the human fetus and to the monkey 1I. Gaull, G. E.: Sulfur amino aclds, folate, and DNA: Metabolic interrelationships fetus, there is a decrease in specific activity of methionine during fetal development. Proceedings of the Sir Joseph Barcroft Centenary adenosyltransferase in later stages of gestation (Fig. 4). At term, Symposium (Cambridge University Press. Cambrldge. 1973). when survival for the rabbit is possible, the activity of lung 12. Gaull, G. E., Sturman, J. A,. and R'diha. N. C. R.: Development of mammalian sulfur metabolism: Absence of cystathionase in human fetal tissue. Pediat. methionine adenosyltransferase also increases. Res., 6: 538 (1972). There are two pathways for synthesis of phosphatidylcholine 13. Gaull, G. E.. Sturman, J. A,, and Rassin, D. K.: Enzymatic and metabolic studies (lecithin): (I) the choline pathway which requires preformed of homocystinuria: Effects of pyridoxine. Neuropadiatrie, 1. 199 (1969). choline and which has not been shown to require S-adenosylmeth- 14. Gaull, G. E., von Berg, W.. Raiha, N. C. R., and Sturman. J. A,: Development of ionine; (2) the methylation pathway in which 3 molecules of methyltransferase activities of human fetal tissues. Pediat. Res., 7: 527 (1973). 15. Gluck. L.. Chez, R., Kulovich, M. V.. Hutchinson, D. L., and Niemann. W. H.: S-adenosylmethionine are required for each molecule of phos- Comparison of phospholipid indicators of fetal lung maturity in the amniotic phatidylcholine synthesized (38). Gluck and coworkers (16-19) fluid of ihr monkey (rMaiaca n~ulaiia)and baboon (Papio papio). Amer. J. have provided indirect data in support of their hypothesis that in Obstet. Gynecol., 120: 524 (1974). the human fetus from 22 to 35 weeks of gestation it is the 16. Gluck, L., Kulovich, M., Eidelman, A. I., Cordero, L., and Khazin, A. F.: Biochemical development of surface activity in mammalian lung. IV. Pulmo- methylation pathway that is involved in the de novo synthesis of nary lecithin synthesis in the human fetus and newborn and etiology of the surfactant lecithin (phosphatidylcholine); beginning at 35 weeks respiratory distress syndrome. Pediat. Res., 6: 81 (1972). the choline pathway becomes quantitatively important. There is 17. Gluck, L.. Landowne, R. T., and Kulovich, M.: Biochemical development of considerable disagreement about the importance of the methyla- surface activity in mammalian lung. 111. Structural changes in lung lecithin during development of the rabbit fetus and newborn. Pediat. Res., 4: 353 tion pathway (5, 38). The present results are consistent with the (1970). proposed existence of the methylation pathway in human and 18. Gluck, L., Motoyama, E. K., Smits, H. L., and Kulovich, M. V.: The monkey fetal lung, although they do not establish its existence or biochemical development of surface activity in mammalian lung. 1. The its quantitative significance. The recent work of Salerno and surface active phospholipids; the separation and distribution of surface-active lecithin in the lung of developing rabbit fetus. Pediat. Res., 1: 237 (1967). Beeler (27) raises the possibility that S-adenosylmethionine also 19. Gluck, L.. Schribney, M., and Kulovich, V.: The biochemical development of may play a role in the choline pathway for the synthesis of surface activity in mammalian lung: The biosynthesis of phospholipids in the phosphatidylcholine. S-Adenosylmethionine has been known to be lung of the developing rabbit fetus and newborn. Pediat. Res.. 1: 247 (1967). the methyl donor during the methylation of phosphatidylethanola- 20. Hancock, R. L.: S-Adenosylmethionine-synthesizing activity of normal and neoplastic mouse tissues. Cancer Res.. 26: 2425 (1966-1). mine to phosphatidylcholine ("methylation pathway") (1, 39). 21. Lowry, 0.H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.: Protein Salerno and Beeler showed rapid incorporation of [rne~hyl-~H] measurement with the Folin phenol reagent. J. Biol. Chem.. 193: 265 (1951). 550 FARRIAUX AND DHONDT

22. Mudd, S. H., Finkelstein, J. D., Irreverre. F., and Laster. L.: Transsulfuration in methionine-activating enzyme. Fed. Proc., 31: 576 (1973). mammals: Microassays and tissue distribution of three enzymes of the 35. Tallan, H. H.. and Gaull. G. E.: Unpublished data. pathway. J. Biol. Chem., 240: 4382 (1965). 36. Tanimura, T., Nelson. T.. Hollingworth, R. R., and Shepard. T. H.: Weight 23. Natori, Y.: Studies on ethionine. VI. Sex-dependent behavior of methionine and standards for organs from early human fetuses. Anat. Rec., 171: 227 (1971). ethionine in rats. J. Biol. Chem., 238: 2075 (1963). 37. Taylor. R. T., and Weissbach, H.: Radioactive assay for serine transhydrox- 24. Niemann, W.: Unpublished observations. ymethylase. Anal. Biochem., 13: 80 (1965). 25. Pascal, T. A., Gillam, B. M., and Gaull, G. E.: Cystathionase: lmmunochemical 38. Villee, C. A,. and Villee, D. B.: Respiratory Distress Syndrome Conference evidence for absence from human fetal liver. Pediat. Res., 6: 773 (1972). Proceedings (Academic Press. New York. 1973). 26. Rose, W. C., and Wixom, R. L.: Amino acid requirementsof man. XIII. Sparing 39. Wilson, J. D., Gibson, K. D., and Udenfriend, S.: Studles on the precursors of effect of cystine on the methionine requirement. J. Biol. Chem., 216: 95 (1955). methyl groups of choline in rat liver. J. Biol. Chem.. 235: 3213 (1960). 27. Salerno, D. M., and Beeler, D. A,: The biosynthesis of phospholipids and their 40. Zachmann, R. D.: The enzymes of lecithin biosynthesis in human newborn lungs. precursors in rat liver involving de novo methylation, and base-exchange 1I.Methionine-activating enzyme and phophat~dyl methyltransferase. Biol. pathways, in vivo. Biochim. Biophys. Acta, 326: 325 (1973). Neonate, 20; 448 (1972). 28. Sturman, J. A.. and Gaull, G. E.: Polyamine biosynthesis in human fetal liver and 41. We are grateful to Dr. Wendell Niemann for determination of gestational age in brain. Pediat. Res., 8: 321 (1974). the monkeys. Mrs. Susan Sansevero gave expert technical assistance. 29. Sturman, J. A,. and Gaull, G. E.: Unpublished studies. 42. Dr. Sternowsky, on leave from the University of Hamburg. Pediatric Clinic, was 30. Sturman, J. A., Gaull, G., and Raihl. N. C. R.: Absence of cystationase in the recipient of a grant from the Deutsche ForschungsgemeinschaR. Bad human fetal liver: Is cystine essential'? Science, 169: 74 (1970). Godesberg. 31. Sturman, J. A,, Gaull, G. E., and Rzihl, N. C. R.: DNA synthesis from the 43. This study was supported by the New York State Department of Mental Hygiene, @-carbon of serine by fetal and mature human liver. Biol. Neonate, 27: 17 National Institutes of Health Clinical Genetics Center Grant GM-19443, and (1975). the Sigrid Juselius Foundation. 32. Sturman, J. A,. Niemann, W. H., and Gaull, G. E.: Metabolism of 3SS-methio- 44. Requests for reprints should be addressed to: G. Gaull, M.D., Department of nine and 3SS-cystine in the pregnant Rhesus monkey. Biol. Neonate, 21: 16 Pediatric Research, New York State Institute for Basic Research in Mental (1973). Retardation, 1050 Forest Hill Rd., Staten Island, N.Y. 10314 (USA). 33. Tallan, H. H.: Unpublished observations. 45. Act epted for publication December 3, 1975. 34. Tallan. H. H., Cohen, P. A,. and Gaull, G. E.: Allosteric properties of rat liver

Pediat. Res. 10: 550-552 (1976) Glycine intestine iminoacids type I hyperprolinemia

Type I Hyperprolinemia: A Study of the Intestinal Absorption of Proline, Hydroxyproline, and Glycine

J. P. FARRIAUX"" AND J. L. DHONDT Laboratoire de Recherche, Clinique Pidiatrique, Cite Hospitaliire, Lille, France

Extract similar to that observed in the kidney. However, the inhibition of intestinal iminoacids and glycine transport seems to be due to Intestinal absorption of proline, hydroxyproline, and giycine was mechanisms more complex than that of a simple inhibition by interpreted by investigation of a type I hyperprolinemia patient and proline. six control subjects. Intestinal perfusion was performed. When proline (Pro), hydroxyproline (OH-Pro), and glycine (Gly) were infused together, an increase in proline concentration did not The mechanisms of membrane transport of proline, hydroxypro- alter aminoacid uptake in the control subjects; however, in the line, and glycine have been investigated in some recent studies, the hyperprolinemia patient, uptake of aminoacids became negligible results of which seem to indicate the presence, in vivo, of a single (Pro, 17-6 pM/min; OH-Pro, 15-0.3 pM/min; and Gly, 13.5-0 transport system for iminoacids and glycine (7, 10, 11, 13). pM/min). Investigation of a type I hyperprolinemia patient allowed us to When each aminoacid was infused alone at increasing concentra- suggest a new approach to the problems of intestinal absorption of tions aminoacid uptake increased in controls; in the hyperproline- these three aminoacids. mic patient, intestinal absorption was less for glycine and hydroxyproline but aminoacid uptake increased with substrate concentra- EXPERIMENTAL PROCEDURE tion; however, for proline, the uptake remained constant (16.5-17 pM/min/ZO cm of intestinal test segment) (Table 1). SUBJECTS When hydroxyproline was infused with an increased concentration of proline in the hyperprolinemic patient, hydroxyproline The subjects were one patient with type I hyperprolinemia uptake first increased (9.8-14.3 pM/min/20 cm) then decreased to whose case is described in an earlier report (4) and six control sub- its basal value, whereas, in the control subjects, uptake increased jects of the same age and weight as the propositus; none had evi- without decreasing subsequently. dence of small bowel or metabolic diseases and plasma aminoacid levels (Pro, Gly, OH-Pro) were normal. Speculation METHODS The chronic hyperprolinemia state might entail adaptation of the transport mechanism with the three infused aminoacids (Pro, A method of intestinal perfusion with which we were already OH-Pro, and Cly), bringing about an "overflow" of the system familiar was adopted (3).